Summary of Work: This research is focused on in vivo approaches to correlate use of alternative/natural environmental species with laboratory animal and cell culture models in the assessment of mutagenic hazard from exposure to chemical or physical agents in the environment. Transgenic technology has been applied to the study of induced somatic mutation directly at the DNA level using PhiX174 bacteriophage as an identical trangenic marker in rodents, fish and cultured cells where dose, adduction, DNA repair, and mutation can be compared. In recent experiments we have determined that spontaneous mutation frequencies in a nontranscribed bacteriophage transgene are similar in mice, fish, and cultured cells and that mutation in the target sequence can be induced in somatic tissues of fish and mice by exposure to an alkylating agent. The effects of exposure to the potent carcinogen 7,12-dimethylbenz[a]anthracene (DMBA) were also investigated in mouse and Fundulus heteroclitus through the analysis of cytochrome P4501A induction and DMBA metabolite levels after i.p. injection with doses of 0.26, 1.9 and 10 and 19 mg/kg in corn oil. Fish showed a linear correlation between mutation in the transgene and CYP1A induction measured as activity (EROD) and protein. The 1.9 and 19 mg/kg doses of DMBA resulted in a 2 and 11-fold increase, respectively, in mutation frequency over controls for the fish. In contrast, the induction in CYP1A activity in mice was much less although, the same metabolites were detected as in the fish with the exception of 7,12-bis- hydroxymethylbenz[a]anthracene. In vivo induced mutation in the transgene was also examined in mice and there was only a 2-fold increase in mutation frequency at the highest dose, similar to the effects observed in fish at the lowest dose. The comparative effects of DMBA indicate the importance of differences in response between sentinel species when evaluating the effects of polycyclic aromatic hydrocarbons. These studies provide evidence that identical gene indicators for specific classes of chemical-induced mutation combined with analysis of biotransformation may provide a mechanistic basis for correlations between laboratory rodents and species that may serve as environmental sentinels in polluted ecosystems. Fish exposed to complex mixtures from known contaminated ecosystems are in the process of being analyzed now. Based on the results here, (1) the target sequence in the transgenic vector will be expanded, (2) mutagenicity in fish exposed to certain drinking water disinfection by-products will be investigated and (3) the approach will be combined with developmental and endocrine disruptor endpoints and applied to assessing toxicities of aquatic mixtures.